Involvement of tropomyosin in the sensitivity of Na+ + K+ ATPase to ouabain

BiochemicalPharmacology,Vol. 29, pp. 297-300. @ Pergamon Press Ltd.1980. Printed inGreat Britain.

OCO6.2952180/02Ol-O297 $02.00/O

INVOLVEMENT OF TROPOMYOSIN IN THE SENSITIVITY OF Na++K+ ATPase TO OUABAIN DAN&ECHARLEMAGNE,* ALAIN

JOCELYNE LEGER,t KETTYSCHWARTZ,?BLANDINEGENY,* ZACHOWSKI* and LIONEL LELIEVRE$,§

Station de Recherches de Virologie et d’Immunologie de l’I.N.R.A., 78850 Thiverval-Grignon, France t1.N.S.E.R.M. (U. 127), Unite de Recherche sur le metabolisme du coeur et des vaisseaux, Hopital *E.R.A.

No. 802 C. N. R. S.,

Lariboisiere, 41 Bd de la Chapelle, 75010 Paris, France $Institut de Recherches en Biologie Moleculaire-Unite des Biomembranes, C.N.R.S., Jussieu 75005 Paris, France

2 Place

(Received 23 March 1979; accepted 27 June 1979) Abstract-

The Na+/K+ ATPase sensitivity to ouabain was shown to be increased by 300 to lOOO-fold after treatment of the plasma membrane by EDTA. Addition of proteins detached from the plasma membrane with Ca2+ ions to EDTA treated membranes reconstituted the original Nat/K+ AVase

resistance to ouabain inhibition. Tropomyosin with Ca2+ ions (not with Mg2+ ions) induced the same effect. When suboptimal doses of tropomyosin were used for such a reconstitution, the dose-response curve indicated a full reconstitution of a given percentage of enzyme molecules. This observation led us to assume a direct or indirect effect of tropomyosin on Na+/K* ATPase functions.

Cardiotonic steroids react with and inhibit the (Na+ + K+) stimulated Mg2+ ATPase (Na+ +K+ ATPase E.C.3.6.1.3.) [l] in membranes. A remarkable feature of the inhibitory effect of ouabain is the difference in sensitivity to this inhibitor between species: as a general rule, rodent cells [2-4] are much more resistant to the drug than human [5] and pig [3] cells are. In some experiments on selected ouabain resistant cell lines, obtained by mutagenic agents, the difference in ouabain sensitivity of the enzyme has been related to mutation at the enzyme level [2, 5, 61. However, from murine plasmocytoma MOPC 173, ouabain resistant variant cell lines are characterized by an increased resistance to CAMP, Concanavalin A and theophylline [7], but with almost the same number of ouabain binding sites as in the wild type cells [8, 91. Furthermore, the sensitivity to ouabain of a plasmocytoma wild type MFzS was shown to be modified by proteins removed by EDTA treatments from the plasma membrane [lo]. Indeed, we have shown these peripheral proteins to be located at the inner face of the plasma membrane and removed from insideout vesicles leading to a 300 fold increase of the Na+ + K+ ATPase sensitivity to ouabain [ 111; moreover, such proteins, after being extensively dialysed against EDTA free medium, led to a complete recovery of the original resistance of the enzyme to ouabain, when added back to EDTA treated insideout vesicles plus Ca2+ ions . [ll]. Cell movements and other functions involved in the transfer of information through the plasma membrane were shown to be linked in some way to the cytoskeleton [12, 131. Thus we put forward the 5 Present address: c/o Dr. Schwartz, Department of Pharmacology and Cell Biophysics, University of Cincinnati, Medical Center, 231 Bethesda Avenue, C, OH 45267, U.S.A. 297

suggestion that proteins belonging to the microfilaments and/or microtubules could be present in the EDTA treated membrane supernatant and involved in the shift of the ouabain sensitivity as described above [lo, 111. Proteins belonging to the microfilaments such as actin, myosin, tropomysin and (Y actinin have been shown to retain similar structure and functions through the different species and organs [ 14-161. These observations prompted us to check the possible effect of four rabbit muscle proteins on the sensitivity to ouabain of EDTA treated plasmamembrane bound Na+ + K+ ATPase. Actin (mol. wt 43,000 daltons), troponin C (mol. wt 18,000 daltons), troponin T (mol. wt 37,000 daltons) and tropomyosin (mol. wt 34,000 and 36,000 daltons) were used. These proteins were chosen either because they bind Ca2’ or exhibit molecular weights close to those of the active proteins in the EDTA supernatant (mol. wt 30,000 - 40,000 daltons). In the following presentation, we demonstrate that only tropomyosin is able to modify the sensitivity of the Nat + K+ ATPase to ouabain in the EDTA treated plasma membranes. MATERIALSANDMETHODS Cells. MFzS cells derived from the MOPC 173 murine plasmocytoma grown as ascites in mice [17] were used. Plasma membrane purification. Cells were lysed in a hypotonic medium and plasma membranes were purified on a discontinuous sucrose gradient and tested for purity as previously described [ 181. In fact, these plasma membranes are mixtures of right sideout (RSO) and inside-out (IO) vesicles. RSO were shown to be permeable to ATP and IO to ouabain but both were impermeable to proteins [ll]. Characterization of these vesicles have been previously published [19]. Protein composition, determined by

298

D. CHARLEMAGNE et al.

SDS polyacrylamide slab gel electrophoresis, has been shown to be the same for IO and RSO vesicles. EDTA

treatment of purified plasma membranes.

Plasma membranes (0.5 mg/ml proteins) were incubated for 15 min with magnetic stirring at 4” in a pH6.8 buffer containing 250mM sucrose, 30mM Imidazole chloride (IS buffer) and centrifuged for 30 min at 31,OOOg. The pellet was suspended (0.5 mg/ml proteins) in ISE buffer (1 mM EDTA containing IS buffer) and incubated for 60 min at 4“ with magnetic stirring. After centrifugation as above, the pellet was again suspended in ISE buffer and submitted to the same treatment. After centrifugation, the pellet suspended in 60 per cent glycerol was stored at -20”. In all cases, centrifugation of EDTA treated membranes was necessary to affect the ouabain sensitivity

WI.

Assay of Na+ + K+ ATPase. The activity was assayed according to Ottolenghi [20]. Kinetic studies (0 - 10 min) were performed by transfering aliquots at given times to test-tubes where reaction was stopped by an acidic molybdate solution and inorganic phosphate was determined according to Anner and Moosmayer [21]. The ATPase activity inhibited by 1 mM ouabain was similar to (Na+ + K+) stimulation of ATP hydrolytic activity of the membranes. Thus we can assume that it represents the Na+ + K+ ATPase activity. The ouabain sensitivity of Nat + K+ ATPase activity was tested with 1O-8-1O-3 M ouabain (Calbiochem) prepared daily. The mixture (membrane + ouabain) was pre-incubated for 10min at 37” in the assay medium. The reaction was initiated with the addition of prewarmed ATP and carried out over 6 points to assess linearity (aliquots taken every 2 min during a period of 10 min). We excluded the possibility that inhibitory activity of ouabain is greatly underestimated at low concentration due to a lag time in the binding of the drug since either a 10 min or a 70 min assay in the presence of varying doses of the drug led to the same inhibition of the enzymatic activity. Furthermore, the amount of Pi liberated over these assays was linear with respect to time, suggesting that enzyme-ouabain equilibrium was reached in less than 2 min (the shortest time used) and that ouabain formed a reversible complex with Na+ + K+ ATPase as shown by Allen and Schwartz [22] with preparations from rat heart (insensitive species).

Fig. 1. Dose-response

vs ouabain membranes

curve of Na++K+ ATPase activity concentration (logarithmic scale) in native (U) and EDTA treated membranes

(A-A).

The 50 per cent inhibition by ouabain of Na++K+ ATPase activity was symbolized as E l/2 [23]. Figure 1 represents the ouabain sensitivity of Na+ + K’ ATPase activity in original membranes and in the ISE pellet (membranes treated successively 1.5min by IS and twice 60 min by ISE). Reconstitution experiments. EDTA treated membranes (150 pg) were incubated for 30-90 min, at 4”, in 1 ml IS buffer plus CaCh(0.1 mM) either with 15 ,ug of dialysed EDTA supernatant proteins or with 10-20 pg of purified proteins, as stated below. It was shown that reconstitution was only possible with calcium (0.01-0.1 mM) and not magnesium ions and that Ca*+, in our experimental conditions, did not inhibit the enzyme nor modify its sensitivity to ouabain [24]. Each experiment was performed several times on 3-5 membrane preparations and 3 different protein preparations. Purifiedproteins. All proteins were obtained from rabbit white muscle and checked for purity in polyacrylamide slab gel electrophoresis according to Laemmli [25]. Actin was prepared according to Spudich and Watt [26], tropomyosin, troponin C and troponin T according to Harstshorne and Mueller [27]. Protein concentration was determined by the method of Lowry et al. [28]. RESULTS

(1) Modification of Na+ + K+ ATPase sensitivity to ouabain by EDTA treatment. In native plasma membranes, the Na++K+ ATPase activity was found to be 50 per cent inhibited by 120 2 20 PM ouabain (symbolized as E l/2). After EDTA treatment, these plasma membranes, which are mixtures of right-sideout (RSO) and inside-out (IO) vesicles, exhibited a two step dose-response curve to ouabain; the first step is correlated to inhibition by ouabain of Na++K+ ATPase activity of IO vesicles : the E l/2 (0.1-0.3 PM) of this Na++K+ ATPase population was 300 to lOOO-fold lower than that of the native membranes. The second step reflects the inhibition of RSO vesicles (El/2 = 120 t 20 PM) (Fig. 1) [ll]. Furthermore, EDTA treatment removes a considerable fraction of the membraneous proteins. In case of RSO-vesicles, the protective proteins remain caught intravesicularly, thus keeping the ATPase insensitive against ouabain. In contrast the IO-vesicles will be depleted of the protective proteins by dilution, thus becoming more sensitive to ouabain. (2) Recovery of the original sensitivity of Na’+K+ ATPase to ouabain. It was shown previously that addition of 15 pg of proteins from the dialysed supernatants of EDTA-treated membranes and of 0.1 mM Ca2+ to 1.50pg of EDTA treated membranes led to the reconstitution of the original Na++K+ ATPase sensitivity to ouabain [lo]. Thus, in a first set of experiments, 150 pg of EDTA treated membranes were incubated in the presence of 0.1 mM Ca2+ with 10-20 pg of one of the following purified proteins: actin, troponin C, troponin T and tropomyosin. It can be seen in Fig. 2 that none of them was active except tropomyosin. While the El/2 for the first step was found to be 0.1-0.3 PM for the EDTA treated membranes either alone or incubated

299

Sensitivity of Na++K+ ATPase to ouabain

00

10“

-(I

-s

IO-' Ouab:: Conc.nt::tlon(M)

lo“

Fig. 2. Dose-response curve of Naf+Kf ATPase activity vs ouabain concentration (logarithmic scale) in EDTAtreated membranes (150 pg) + 0.1 mM CaClz + tropomyosin (10 or 2Opg (0--O); and EDTA-treated membranes (150 pg) + 0.1 mM CaCl + either actin or troponin C or troponin T (10 or 20 pg) (B---Cl).

with actin,

troponin C, troponin T or Ca2+ ions alone, it increased to 120 + 20 PM with tropomyosin.

In a second set of experiments, 150 wg of EDTA treated membranes were incubated with 10 Fg of tropomyosin and either no ions or 0.1 mM Ca2’ ions or 1 mM Mg+ ions. It can be seen in Fig. 3 that only tropomyosin with Ca2’ ions was able to restore the original sensitivity of the enzyme to ouabain, while Mg2+ ions are inactive in this phenomenon. In few cases, tropomyosin alone modified the E l/2 from 0.3 PM up to 2 FM, probably due to the presence of Ca” ions in the tropomyosin solution, since replacement of IS buffer by ISE buffer in the incubation suppressed this partial modification. (3) Stepwise reconstitution of the ouabain binding site of the Na’+K+ ATPase. In order to determine the reconstitution mechanisms induced by tropomyosin, we looked for the restoration capacities of increasing amounts of tropomyosin. It can be seen on Fig. 4 that (i) amount of tropomyosin up to 0.075 pg are unable to induce any shift of the sensitivity of the enzyme to ouabain; (ii) doses of tropomyosin from 0.3 to 0.6 gg can modify the shape of the dose-response curve of Na++K+ ATPase to ouabain with a stepwise increasing height of the plateau; (iii) doses of tropomyosin from 1 to 20 kg lead to a complete restoration of the sensitivity of the enzyme to ouabain identical to original plasma membranes or to EDTA treated membranes reconstituted with Ca*’ ions by proteins from ISE supernatant.

Fig. 3. Dose-response curve of Na++K+ ATPase activity vs ouabain concentration (logarithmic scale) in EDTAtreated membranes (150 pg) + 0.1 mM CaCl2 + tropomyosin (10 or 20 pg) (C--O); and EDTA-treated membranes (150 Fg) + tropomyosin (10 or 20 pg) either with 1mM MgClz or without CaClz in ISE buffer (m-m). B.P.29/3

r

Fig. 4. Dose-response curve of Na’/K+ ATPase activity vs ouabain concentration (logarithmic scale) in EDTAtreated membranes (150 pg) + 0.1 mM CaCl2 + tropomyosin (l-10 pg) (O--0); EDTA-treated membranes (150 pg) + 0.1 mM CaCl2 + tropomyosin (0.3 0.6 pg) (u); and EDTA-treated membranes (150 pg) + 0.1 mM CaCl2 + tropomyosin (0.075 pg) (V-V).

During stepwise reconstitution, note that the E II2 values of both steps are never modified (0.3 PM and 120 PM). DISCUSSION

susceptibility of the Na++K+ ATPase to ouabain can be drastically increased by an EDTA treatment of purified plasma membranes. The total activity of the enzyme was unaffected by this treatment, in spite of the significant loss (30-40 per cent) of membrane associated proteins [lo]. We have shown that the original resistance of the enzyme was recovered by addition of proteins from ISE supernatant together with Ca2+. This was also obtained by addition with Ca’+ of rabbit muscle tropomyosin, while actin, troponin C and troponin T were revealed to be inactive even at higher doses. The stepwise reconstitution suggests that there are two populations of ATPase exhibiting a different sensitivity to ouabain (E 112 = 0.3 FM and 120 PM respectively), the sensitive one being shifted by tropomyosin to a resistant one (without any intermediate step). As proteins from ISE supernatant act only on IO vesicles [ll], tropomyosin might be assumed to act at the inner face of the plasma membrane to induce the recovery of the original resistance of the enzyme Thus, the ouabain binding sites to ouabain. expressed at the external face would be modified by events occuring at the inner face. As yet we know nothing about the structure of the reconstituted ouabain binding sites, whether it is similar or mimics the original structure. Until now, tropomyosin was known to bind actin in the microfilaments, but has never been described as a component of the plasma membrane. It is also important to stress the point that tropomyosins are not recognized to be Ca*+ binding proteins while here, Ca2+ ions are required for the shift experiments and Mg2+ cannot substitute Ca*+: thus we might hypothetise that there is no direct interaction between Ca2+ and tropomyosin and that the way Ca2+ is active on the reconstitution between tropomyosin and plasma membranes is indirect. Several membrane bound enzymes have already been shown to be modulated by proteins which can be removed by EDTA and/or are assumed to belong The

300

D. CHARLEMAGNE

to the cytoskeleton: Ca’++M$+ ATPase activator protein can be released by EDTA treatment of human erythrocyte plasma membranes and reconstitute the EDTA treated plasma membranes [29]. The ornithine decarboxylase activity is inhibited by colchicine and vinblastine [30], suggesting a regulatory mechanism of the enzyme linked to the integrity of the cytoplasmic microtubules. Taken together with our data reported here and elsewhere [ 111, these results led us to conclude that proteins located at the inner face of the plasma membrane or belonging to the cytoskeleton may be modulators of membrane bound enzymes and modifiers of specific binding sites located at the external face of the plasma membrane. The apparent discrepancy between the two steps of ouabain binding on intact cells and the homogeneity of the dose-response curve of the Na++K+ ATPase vs increasing amounts of ouabain in purified plasma membranes is due to the different materials used: in cells the cytoskeleton is intact while in plasma membranes microfilaments disappeared. Thus if the two step phenomenon is due to an interaction between the enzyme and a specific contractile protein of the intact cytoskeleton as suggested by our experiments, it is unlikely to be observed with

plasma membranes. In intact cells, ouabain binding will at first stimulate a transmembrane signal leading to a change in the cytoskeleton structure, which in turn could modify the ouabain binding site structure through a change in the relationship between the cytoskeleton and the inner face plasma membrane: thus, we observed a two step binding of ouabain. Acknowledgements-We

would like to thank Dr. A. Paraf for suggestions during the work and criticism of the manuscript, and J. Muller, M. Nezonde, and Y. Fedon for their technical assistance. This work was supported by la Delegation GCnCrale a la Recherche Scientifique et Technique (contrat 77 7 1267) by 1’Institut National de la Sante et de la Recherche Meditale (contrat 27 76 59) by la Ligue Nationale contre le Cancer and by la Fondation pour la Recherche Medicale Fran9aise. REFERENCES 1. J. C. Skou, Biochim. biophys. Acta 42, 6 (1960) 2. R. M. Baker, D. M. Brunette, R. Mankovitz, L. H. Thompson, G. F. Whitmore, L. Siminovitch and J. E. Till, Cell 1, 9 (1974).

etal.

3. M.R. Quastel and J. J. Vogeffanger, Cell. Immun. 2, 504 (1971). 4. D. F. Hulser, H. J. Ristow, D. J. Webb, H. Pachowsky and W. Franck, Biochim. biophys. Acta 372,85 (1974). 5. H. M. Rosenberg, J. Cell. Physiol. 85, 135 (1975). 6. R. M. Baker, in Biogenesis and Turnover of Membrane Macromolecules, p. 93. Raven Press, Hewlett (1976).

7. A. Paraf, E. Faivre, V. Zilberfarb, A. Zachowski and L. Lelievre, C. r Acad. Sci. Paris 281, 567 (1975). 8. B. Geny, L. Lelievre, D. Charlemagne and A. Paraf, Expl. Cell Res. 120, 383 (1979). 9. D. Charlemagne, L. Lelievre, P. Jolles and A. Paraf, Biochem. Pharmac. 28, 3513 (1979). 10. L. Lelievre, D. Charlemagne and A. Paraf, Biochem. biophys. Res. Commun. 72, 1526 (1976).

11. A. Zachowski, L. Lelievre, J. Aubry, D: Charlemagne and A. Paraf, Proc. natn. Acad. Sci. U.S.A. 74, 633 (1977). 12. T. D. Pollard and R. R. Weighing, CRC Critical Reviews in Biochemistrv Vol. 2. D.1. Chemical Rubber

Company, Cleveland, Ohio (19?4). 13. J. H. Ash, D. Louvard and S. J. Singer, Proc. natn. Acad. Sci. U.S.A. 74, 5584 (1977). 14. A. Lazarides, J. Cell. Biol. 68, 202 (1976). 15. E. D. Korn, Proc. natn. Acad. Sci. U.S.A. 75, 588 (1978). 16. R. E. Fine and A. L. Blitz, J. molec. Biol. 95, 447

(1975). 17. A. Zachowski and A. Paraf, Biochem. Commun.

biophys. Res.

57, 787 (1974).

18. J. Aubrv. A. Zachowski. A. Paraf and J. Colombani, Arms. Immun. Inst. Pasteur. 130, C, 17 (1979). 19. L. Lelievre. Biochim. bioohvs. Acta. 291. 662 (1973). \ ’ 20. P.’ Ottolenghi, Biochem. j. iS1, 61 (1975). 21. B. Anner and M. Moosmayer, Analyt. Biochem. 65, 305 (1975). 22. J. C. Allen and A. Schwartz, J. Pharmac. exp. Ther. 168, 42 (1969). 23. L. Lelievre, D. Charlemagne and A. Paraf, Biochim. biophys. Acta. 455, 277 (1976). 24. L. Lelievre, A. Zachowski, D. Charlemagne, R. Laget and A. Paraf. Biochim. bioohvs. Acta. 557.399 (1979). ’ 25. U. K. Laemmli, Nature, Lbnd. 227, 680 (1970): 26. J. A. Spudich and S. Watt, .I. biol. Chem. 246, 4866 (1971). 27. D. J. Hartshorne and H. Mueller, Biochim. biophys. Acta 175, 310 (1969). 28. 0. H. Lowry, N. J. Rosebrough, A. L. Farr and R. J. Randall, J. biol. Chem. 193, 265 (1951).

29. D. T. Hanakan, R. D. Taverna, D. D. Flynn and J. E. Ekholm, Biochem. biophys. Res. Commun. &I,1009 (1978). 30. K. Chen, J. Heller, and E. S. Canellakis, biophys. Res. Commun. 68, 401 (1976).

Biochem.